﻿Connective Cooling in l<luids. 



937 



the liquid. The factor ?/t+25/^t is given in Table II. and 

 it is seen that the temperature change is considerable, 

 particularly with glycerine. 



From results obtained, and using the above correction for 

 temperature changes in viscosity, Table V. and fig. 3 have 

 been derived, and each liquid is represented on the graph by 

 four points. The extreme points of each set refer to the 

 extremities of the range of the experimental results. They 

 refer respectively to the cooling of the fine wire at 2° 0. 

 excess and that of the other wire at 50° C. excess. 



Table V. 

 Relation between Hjk and gd?a6c\kv. 



Fluid. 



Airt 



COli 



^nilh 



Wire diain . 



00083 

 00155 



0-0083 

 00155 



0-0083 

 00155 



00083 

 00155 



Glycerine j 00083 



00155 



Olive oil 



2° C. excess. 



50° C. excess. 



g&aQc/kv. 



H/&. 



gd?aQc/kv. 



H/& 



3 120 



•0 3 83 



1-41 



•0 2 235 

 0153 



1-66 

 1-99 



•229 

 1-49 



2-55 

 342 



790 

 513 



3-86 

 562 



0225 

 1-47 



204 

 2-61 



1-21 

 7-9 



3-56 

 4-74 



•0 3 67 



•0 2 44 



1-34 

 1-73 



•047 

 •30 



2-18 

 3-07 



•0 4 87 

 •0 3 56 



1-18 

 1-44 



•0 2 21* 

 0134* 



1-51* 

 1-99* 



* These refer to 9 = 20° C 



t On the graph of fig. 3 are also plotted the following values for the wires 

 at 2U0° C. excess in air : — 



Wire diain. .. 



0-0083 



00155 



gd*aOc/kv .... 



. 5-12x10-* 



3-00 XlO- 2 



T&lk ... 



1-81 



214 



It is seen that the points all lie very well on one curve, so 

 that equation (16) appears to represent the results of the 

 present experiments very satisfactorily. The upward exten- 

 sion of the curve beyond the points plotted is the 

 representation on the present basis of the upper part of the 

 curve for gases already referred to. 



Phil. Mag. S. G. Vol. 44. No. 263. Nov. 1922. 3 P 



